动态信息网络中基于角色的结构演化与预测

动态信息网络是当前复杂网络领域一个极具挑战的新问题,其动态的演化过程具有时序、复杂、多变的特点.结构是网络最基本的特征,也是进行网络建模和分析的基础,研究网络结构的演化过程对全面认识复杂系统的行为倾向具有重要意义.使用“角色”来量化动态网络的结构,得到动态网络的角色模型,应用并改进多类标分类问题的“问题转换”思想,将动态网络的角色预测问题视为多目标回归问题,以历史网络数据作为训练数据构建模型,预测未来时刻网络可能的角色分布情况,提出基于多目标回归思想的动态网络角色预测方法MTR-RP.该方法不仅克服了基于转移矩阵方法忽略时间因素的不足,并且考虑了多个预测目标之间可能存在的依赖关系,实验结果表明,本文提出的MTR-RP方法具有更准确且更稳定的预测效果.

Role-Based Structural Evolution and Prediction in Dynamic Networks

The research of dynamic information networks is a new and very challenging problem in the field of current complex networks.The evolution of dynamic networks is temporal,complex and changeable.Structure is the basic characteristics of the network,and is also the basis of network modeling and analysis.And the study of the network structure evolution is of great importance in getting a comprehensive understanding of the behaviors of complex systems trend.This paper introduces "role" to quantify the structure of dynamic network and proposes a role-based model.To predict the role distributions of dynamic network nodes in future time,this paper views the role prediction as a multi-target regression problem.Extracting properties from historical snapshots sub-network and regarding the future role distributions of dynamic network nodes as the prediction target.And this paper proposes the method of multi-target regression based role prediction of dynamic network (MTR-RP).This method not only overcomes the drawback of existing methods based on transfer matrix which ignore the time factor,also takes into account the possible dependencies between multiple forecast target.Experiments results show that MTR-RP has better and more stable prediction effect compared with the existing methods.